JOURNAL OF CLINICAL MICROBIOLOGY, Mar. 1989, p. 394-399 Vol. 27, No. 30095-1137/89/030394-06$02.00/0Copyright © 1989, American Society for Microbiology

Phialophora repens, an Emerging Agent of SubcutaneousPhaeohyphomycosis in Humans

MASAKI HIRONAGA,lt* KAZUKO NAKANO,2 ICHIRO YOKOYAMA,3 AND JUNICHI KITAJIMA4

Department of Dermatology, Shiga University of Medical Science, Otsu 520-21,1 Division ofDermatology2 andDivision of Orthopedics, Izumi City Hospital, Izumi 594, and Department ofDermatology,

Osaka City University of Medical School, Abeno-ku, Osaka 545,4 Japan

Received 18 August 1988/Accepted 15 November 1988

A 63-year-old Japanese man had phaeohyphomycosis that occurred as a solitary subcutaneous nodule on thedorsal aspect of his left hand. In the nodule there were foci of mixed granulomatous and suppurativeinfiltrations circumscribed by thick fibrous tissue reaction. The foci contained short septate hyphae andoccasionally small rounded aggregates of irregularly branched septate hyphae, both of which were nonpig-mented or rarely weakly pale brown. Fungal culture from the nodule was positive for a dematiaceous mold.The mycologic features of the mold were typical of Phialophora repens. The infection was successfully treatedby excision of the nodule. This is the second reported case of infection due to P. repens.

Phialophora repens is a dematiaceous mold commonlyfound in nature as a cause of bluish gray stain of logs andlumber (5, 6). The mold was not recognized as an opportu-nistic pathogen for humans until 1975, when Meyers et al.(15) first reported a case of mycotic granuloma caused by P.repens. The infection occurred as a subcutaneous granulo-matous nodule in the scalp of an adult man with advancedlepromatous leprosy. Although the lesion did not resemblemycetoma grossly, it was suggestive microscopically by thepresence in some granulomatous foci of small aggregates ofhyphae similar to grains of eumycotic mycetoma. This P.repens infection was redescribed later by Emmons et al. (8)as being a case of phaeomycotic cyst.

Recently we encountered a second case of subcutaneousphaeohyphomycosis caused by P. repens in an adult man.The purpose of this paper is to present the morphologiccharacteristics of this isolate, a clinical pathologic report ofthis case, and a review of the pertinent literature.

CASE REPORT

A 63-year-old Japanese man was admitted to the hospitalon 3 October 1983. Physical examination revealed a 1.3- by1.5-cm elevated, erythematous nodule on the dorsal aspectof his left hand. The patient was otherwise healthy. Thenodule had begun, with no noticed antecedent trauma at thesite, about two months earlier and gradually increased insize. There were no draining sinuses or grains. The routinelaboratory examinations and immunologic tests in the pa-tient disclosed no abnormality except for evidence of milddiabetes mellitus. The nodule was excised for examinationimmediately after the admission, but no microbial culturewas made.Examination of hematoxylin- and eosin-stained tissue

sections of the excised nodule revealed a well-delineatedgranulomatous lesion extending from the deep dermis intothe subcutaneous tissue. The subcutaneous tissue was fo-cally fibrotic, and in the areas of fibrosis there were circum-scribed small foci of mixed granulomatous and suppurative

* Corresponding author.t Present address: Hironaga Dermatology Clinic, Katsube-cho,

Moriyama 524, Japan.

inflammations (Fig. 1). Palisading epithelioid cells, giantcells, and lymphocytes surrounded central microabscessescomposed of polymorphonuclear leukocytes. The microab-scesses contained a few, or rarely numerous, short septatehyphae and occasionally loose hyphal masses, both of whichwere nonpigmented or rarely weakly pale brown. Themasses were rounded aggregates of irregularly branchedseptate hyphae of 2.5 to 5 ,um width often with enlargedterminal cells, measuring 200 to 300 by 80 to 200 p.m (Fig. 2).They looked like the grains of eumycotic mycetomas but didnot manifest the Hoeppli-Sprendore phenomenon sometimesseen in the actinomycotic or eumycotic mycetomas. The freehyphae were 2.5 to 3 ,um wide and were also seen among theepithelioid cells that bordered the granulomatous foci.About 1 month after the surgery, a small similar nodule

recurred within the subcutaneous scar of the last operation,

-~~~~~~ ~ ~ ~ ~ ~ ~ ~-:,

'w~~~~~~~~~,_r

.4> Ar#

FIG. i Granulomatous foci circumscribed by thick fibrous tis-sue reaction and containing small rounded aggregates of hyphae.Bar 200 m.m Periodic acid-Schiff stain

394

1.

on October 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

PHIALOPHORA REPENS AND SUBCUTANEOUS PHAEOHYPHOMYCOSIS 395

sa '>Z,,O*t b

FIG. 2. Rounded aggregates of irregularly branched septate hy-phae seen in the central microabscess of a granulomatous focus.Bar, 20 ,um. Periodic acid-Schiff stain.

so that skin biopsy and fungal culture were performed.Examination of the biopsy specimens revealed a few shorthyphae in the nodule. Also, the biopsy was culture positivefor a dematiaceous mold (SM 3531) identified as P. repens.The organism was identified on the basis of the morphologiccharacteristics described and discussed below.

Oral administration of 7.5 g of flucytosine daily wasstarted in the patient but discontinued 3 months later be-cause only a slight clinical improvement had been noted. TheMIC of flucytosine for the isolated organism was 60 ,ug/mlwhen tested later by the method of Shadomy and Espinel-Ingroff (20). Ultimate cure was achieved with additionalsurgical excision of the recurred nodule.The isolate of P. repens was subcultured under various

conditions and maintained in our mycology laboratory at theDepartment of Dermatology, Shiga University of MedicalScience, as no. SM 3531. The living culture of this isolate hasbeen deposited in the American Type Culture Collection,Rockville, Md., with the accession number ATCC 58115.

MATERIALS AND METHODS

Reference strains were supplied by Michael R. McGinnisat the University of Texas Medical Branch, Galveston. Thestrains and their sources were as follows: P. repens CBS294.39 (UTMB [University of Texas Medical Branch] 227),isolated from pine lumber (used to prepare nomenclaturaltype) (6); P. repens CBS 423.73 (UTMB 184), from mycoticgranuloma (15).The organisms were studied after 1 to 4 weeks of incuba-

tion on potato glucose agar (Difco Laboratories) at 25°C or,if necessary, at 37°C. Slide culture preparations were exam-ined by bright-field and Nomarsky differential interferencecontrast microscopies.

Scanning electron microscopy (SEM) was performed asfollows. Briefly, small squares of sporulating thallus weregently cut out of the colony of the organism examined,including the supporting agar block, after incubation at 25°C.The specimens were fixed in 5% glutaraldehyde at 3°C for 12h and further in 1% osmium tetroxide at 3°C for 3 h,dehydrated in a graded ethanol series, and then washed inincreasing concentrations of amyl acetate in absolute ethanol

FIG. 3. Colony of P. repens SM 3531 (ATCC 58115) on potatodextrose agar, 4 weeks old at 25°C.

and finally in absolute amyl acetate. The materials werequickly dried with a critical point dryer (JEE-4C), coatedwith gold by using an ion IB-3 (Eiko Engineering Ltd.), andexamined with a scanning electron microscope S-500S (Hi-tachi) generator operated at 10 kV.

RESULTSColony morphology. The new isolate SM 3531 (ATCC

58115) grew moderately well at 25°C on potato glucose agar;the colony attained a diameter of 2.8 cm after 1 week, 4.2 cmafter 2 weeks, and 7.2 cm after 4 weeks. The colony was flat,moist, and whitish at first and became brown, dark brown toolivaceous brown, and velvety to more or less cottony later,with tufts of grayish white aerial hyphae (Fig. 3). The reverseside was brown, dark olivaceous brown to nearly black withage. The isolate was able to grow at 37°C, but the growth ratewas reduced by about one-half that at 25°C.SM 3531 was closely similar to the reference strain CBS

423.73 in the color and texture of colonies, although the

N.14i

do

j ~~j.is s

)A

fi!

fyf

,L

FIG. 4. Fasciculated pale brown hyphae with slender phialides.Bar, 10 ,um. Nomarsky differential interference contrast micros-copy. All microscopic photographs are of SM 3531 (ATCC 58115).

VOL. 27, 1989

Ji

r

on October 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

396 HIRONAGA ET AL.

FIG. 5. Branched conidiophores with short phialides. Bar, 10,um. Nomarsky differential interference contrast microscopy.

growth rate was slightly higher in the former. However, thetype strain CBS 294.39, as presently examined, was no

longer typical and formed a rapidly growing, pale brown tograyish brown colony with some farinose aerial hyphae andsimilarly colored reverse.

Microscopic morphology. The new isolate SM 3531 pro-

duced septate hyphae that were hyaline at first and palebrown later, smooth, 2 to 4 ,um wide, and often fasciculated(Fig. 4). The conidiophores were hyaline to pale brown,smooth, and undifferentiated. The conidiogenous cells were

intercalary, hyaline to pale brown, smooth, 6 to 30 ,um long,and 1.2 to 3.2 ,um wide just above the base; they were

monophialidic or sometimes polyphialidic and lageniform toelongate and apically had short inconspicuous or slightlyflared collarettes of 1 to 1.8 ,um width (Fig. 4 through 8).Coiled hyphae with phialides were typically formed. SEM ofthe conidiogenous cells revealed an enteroblastic phialidicmanner of conidial development in this isolate. The lageni-form or cylindrical phialides ended in short tubular or

slightly flared collarettes without constriction at the neck;the secondary conidial initials appeared to emerge in basi-

FIG. 6. Slender phialides with narrow short inconspicuous coll-arette and conidia. Bar, 10 pum. Nomarsky differential interferencecontrast microscopy.

FIG. 7. Lageniform phialide with a slightly flared short collaretteand conidia. Bar, 5 ,um. Nomarsky differential interference contrastmicroscopy.

petal succession through the collarettes (Fig. 9 through 11).The phialoconidia were hyaline, smooth, nonseptate, cylin-drical to allantoid, and 3 to 6 by 1 to 2.2 ,um (average, 4.6 to1.6 ,um) (Fig. 7, 8, and 12).

In the microscopic morphology the two reference strainsand our new isolate were almost alike, excluding a minimaldifference in the range of conidial size. Typically penicillatebushes of phialides were not produced in any cultures of thethree strains presently examined.On the basis of the macroscopic and microscopic charac-

teristics described above, we identified our new isolate SM3531 (ATCC 58115) as conspecific with P. repens.

DISCUSSION

P. repens was described as a new species of the genusCadophora Lagerberg by Davidson in 1935 (6). The funguswas later placed in the genus Phialophora Medlar by Conant(5) and subsequently discussed in reviews written by variousworkers (3, 4, 7, 12, 18). Typically penicillate bushes ofphialides were originally described in the type strain (CBS294.39) (6) and have been considered characteristic of P.

FIG. 8. Phialide with two conidiogenous loci. Bar, 5 ,um. Nomar-sky differential interference contrast microscopy.

J. CLIN. MICROBIOL.

on October 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

PHIALOPHORA REPENS AND SUBCUTANEOUS PHAEOHYPHOMYCOSIS 397

om.j _|miFIG. 10. Elongate phialide with a slightly flared collarette and a

conidial initial. Bar, 3 ,um. SEM.

FIG. 9. Rather elongate phialide with a short tubular inconspic-uous collarette and a conidial initial. Bar, 3 ,um. SEM.

repens (4, 7, 18). However, we could find no penicillateconidiophores with phialides in our new isolate SM 3531(ATCC 58115) or in the type strain. As pointed out by deHoog (7), whether this branching pattern has taxonomicsignificance still needs examination of additional isolates.The microscopic morphology of SM 3531 was otherwiseidentical with that of the type strain and of the isolate ofMeyers et al. (CBS 423.73) (15).Some species of Phialophora Medlar or Lecythophora

Nannfeldt (9) microscopically mimic P. repens in somedegree, so that identifying isolates of these species is ratherdifficult (3, 4, 12, 21). The difficulty is also due to the fact thatthe phialides formed by these species are often variable inshape and size in different strains of each species or evenwith the ages of cultures.The closest species to P. repens is Phialophora parasitica,

although as a rule P. repens produces more delicate conid-iophores and phialides than the latter. P. parasitica ischaracterized microscopically by the encrustations of hy-phae and conidiophores, the collarettes with a somewhatconstricted neck, and the sometimes proliferating phialides;in contrast, P. repens lacks all of these characteristics (9, 12,21). Since our new isolate SM 3531 (ATCC 58115) also lacksall of such microscopic characteristics, it does not appear tobe conspecific with P. parasitica.Lecythophora hoffmannii and L. mutabilis are also some-

what close to P. repens but can be distinguished by theFIG. 11. Lageniform phialide with a conidial initial emerging

through its collarette. Bar, 3 ,um. SEM.

VOL. 27, 1989

on October 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

398 HIRONAGA ET AL.

FIG. 12. Cylindrical to allantoid, smooth conidia. Bar, 5 ,um.SEM.

slimy, pink to olivaceous brown colonies and by the predom-inantly forming adelophialides with reduced collarettes. L.mutabilis is also distinct by the centrally brown colonies thatare due to chlamydospores (7, 9, 12, 18). Our isolate SM3531 is also not compatible with these two because of thedifferences in colony and microscopic morphology.Meyers et al. (15) were the first to report P. repens

infection in humans. In the report, they clearly noted thatthere were hyphae and occasionally microcolonies of anonpigmented fungus in tissue sections. Nevertheless, Em-mons et al. (8) described later that the fungal elements wereoften nonpigmented and listed P. repens as an agent ofphaeomycotic cyst. Likewise, most medical mycologistshave regarded P. repens as an agent of phaeohyphomycosis(1, 12, 17). Our histologic observations described herein alsoafford evidence that the hyphal walls of P. repens are rarelyor partially dematiaceous in the human subcutaneous tissue.

In some cases of phaeohyphomycosis, the dematiaceousnature of the fungal elements is not apparent in tissuesections and is revealed only later in culture (17). As in ourcase, however, careful examination of hematoxylin- andeosin-stained tissue sections or unstained sections oftenreveals a few hyphae that are dematiaceous (1). Moreover,the pigment of hyphal walls of a fungus may be variable intissue of different organs as well as even in its differentstrains. In fact, Bipolaris spicifera was seen as brown-pigmented hyphae in the skin or the cornea (13, 23) butformed nonpigmented hyphae when it invaded the brain (22).The inflammatory reaction elicited by P. repens is a mixed

granulomatous and suppurative one and does not differ fromthat in cases of phaeohyphomycosis caused by P. parasitica(2), Phialophora richardsiae (19), Exophialajeanselmei (10),E. moniliae (11, 14), E. spinifera (16), or other dematiaceous

molds (17). P. repens infection, however, can be differenti-ated histologically by the unique aggregates of hyphae intissue sections. Phaeohyphomycosis caused by P. repensthus appears somewhat atypical and intermediate betweenphaeohyphomycosis and eumycotic mycetoma.Our patient had P. repens infection despite the absence of

predisposing factor except for mild diabetes mellitus. Therole of the diabetes mellitus for establishing the fungusinfection in the patient seems minor. In general, phaeohy-phomycosis due to P. repens, whether in a normal host or inan immunologically compromised host (15), is relativelybenign and good in prognosis.

ACKNOWLEDGMENTS

We thank Michael R. McGinnis, Department of Pathology, Uni-versity of Texas Medical Branch, Galveston, for providing authenticand reference strains and the late Tomomasa Fujigaki, Departmentof Dermatology, Shiga University of Medical Science, Otsu, Japan,for technical assistance.

LITERATURE CITED

1. Ajello, L. 1986. Hyalohyphomycosis and phaeohyphomycosis;two global disease entities of public health importance. Eur. J.Epidemiol. 2:243-251.

2. Ajello, L., and L. K. Georg. 1974. A case of phaeohyphomycosiscaused by a new species of Phialophora. Mycologia 66:490-498.

3. Beyma, F. H. 1943. Beschreibung der im Gentraalbueau voorSchimmelcultures vorhandenen Arten der Gattungen Phialo-phora Thaxter and Margarinomyces Laxa, nebst Schlüssel zuihrer Bestimmung. Antonie van Leeuwenhoek J. Microbiol.Serol. 9:51-76.

4. Cole, G. T., and B. Kendrick. 1973. Txanomic studies ofPhialophora. Mycologia 65:661-688.

5. Conant, N. F. 1937. The occurrence of a human pathogenicfungus as a saprophyte in nature. Mycologia 29:597-598.

6. Davidson, R. W. 1935. Fungi causing stain in logs and lumber inthe sourthern states, including five new species. J. Agric. Res.50:789-807.

7. De Hoog, G. S. 1983. On the potentially pathogenic dematia-ceous hyphomycetes, p. 184-193. In D. H. Howard (ed.), Fungipathogenic for human and animals, part A, biology. MarcelDekker, Inc., New York.

8. Emmons, C. W., C. H. Binford, J. P. Utz, and K. J. Kwon-Chung. 1977. Phaeomycotic cyst, p. 425-436. In Medical my-cology, 3rd ed. Lea & Febiger, Philadelphia.

9. Gams, W., and M. R. McGinnis. 1983. Phialemonium, a newanamorph genus intermediate between Phialophora and Acre-monium. Mycologia 75:977-987.

10. Hironaga, M., T. Mochizuki, and S. Watanabe. 1982. Cutaneousphaeohyphomycosis of the sole caused by Exophialajeanselmeiand its susceptibility to amphotericin B, 5-FC and ketoconazole.Mycopathologia 79:101-104.

11. Matsumoto, T., K. Nishimoto, K. Kimura, A. A. Padhye, L.Ajello, and M. R. McGinnis. 1984. Phaeohyphomycosis causedby Exophiala moniliae. J. Med. Vet. Mycol. 22:17-26.

12. McGinnis, M. R. 1978. Human pathogenic species ofExophiala,Phialophora, and Wangiella, p. 37-59. In The black and whiteyeasts. Scientific publication no. 356. Pan American HealthOrganization, Washington, D.C.

13. McGinnis, M. R., M. G. Rinaldi, and R. E. Winn. 1986.Emerging agents of phaeohyphomycosis: pathogenic species ofBipolaris and Exserohilum. J. Clin. Microbiol. 24:250-259.

14. McGinnis, M. R., D. F. Sorrel, R. L. Miller, and G. W.Kaminski. 1981. Subcutaneous phaeohyphomycosis caused byExophiala moniliae. Mycopathologia 73:62-72.

15. Meyers, W. M., J. R. Dooley, and K. J. Kwon-Chung. 1975.Mycotic granuloma caused by Phialophora repens. Am. J. Clin.Pathol. 64:549-555.

16. Nielsen, H. S., and N. F. Conant. 1986. A new human pathogenic

J. CLIN. MICROBIOL.

on October 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from

PHIALOPHORA REPENS AND SUBCUTANEOUS PHAEOHYPHOMYCOSIS 399

Phialophora. J. Med. Vet. Mycol. 6:228-231.17. Rippon, J. W. 1988. Medical mycology. The pathogenic fungi

and the pathogenic actinomycetes, 3rd ed., p. 297-324. TheW. B. Saunders Co., Philadelphia.

18. Schol-Schwarz, M. B. 1970. Revision of the genus Phialophora(Moniliales). Persoonia 6:59-94.

19. Schwartz, I. S., and C. W. Èmmnons. 1968. Subcutaneous cysticgranuloma caused by a fungus of wood pulp (Phialophorarichardsiae). Am. J. Clin. Pathol. 49:500-505.

20. Shadomy, S., and A. Espinel-Ingroff. 1974. Susceptibility testingof antifungal agents, p. 569-574. In E. H. Lennette, E. H.Spaulding, and J. P. Truant (ed.), Manual of clinical microbiol-

ogy, 2nd ed. American Society for Microbiology, Washington,D.C.

21. Weitzman, I., M. A. Gordon, R. W. Henderson, and W. L.Edward. 1984. Phialophora parasitica, an emerging pathogen.J. Med. Vet. Mycol. 22:331-339.

22. Yoshimori, R. N., R. A. Moore, H. H. Itabashi, and D. G.Fujikura. 1982. Phaeohyphomycosis of brain: granulomatousencephalitis caused by Drechslera spicifera. Am. J. Clin.Pathol. 77:363-370.

23. Zapater, R. C., E. J. Albesi, and G. H. Garcia. 1975. Mycotickeratitis by Drechslera spicifera. J. Med. Vet. Mycol. 13:295-298.

VOL. 27, 1989

on October 1, 2020 by guest

http://jcm.asm

.org/D

ownloaded from